Programmed by: Yolanda Larriba, Department of Statistics and Operational Research, University of Valladolid, Valladolid, Spain
Normalization is a key step in the pre-processing of microarray gene-expression data, since it reduces data variability. Yet, it may significantly affect the subsequent analyses, such as the rhythmicity detection. This algorithm introduces a bootstrap based methodology which allows not only to robustly identify rhythmic genes, but also to simulate microarray gene-expressions within oscillatory systems (c.f. circadian clock) from a reference microarray data set.
The implemented R code considers the same seven normalization methods proposed in , namely Quantile (Bolstad et al., 2003), (Cyclic) Loess (Bolstad et al., 2003), Contrast (Astrand, 2003), Constant (Bolstad et al., 2003), Invariant Set (Li and Wong, 2001), Qspline (Workman et al., 2002) and Variance Stabilization Normalization (VSN) (Huber et al., 2002). Rhythmicity detection is also conducted according to the same three rhythmicity detection algorithms considered in that paper, i.e. ORIOS (Larriba et al., 2016), JTK_Cycle (Hughes et al., 2010) and RAIN (Thaben and Westermark, 2014).
Upon downloading the software read the README.pdf for details on how to use the code.
 Larriba, Y., Rueda, C., Fernández, M.A. and Peddada, S.D. (2018). A bootstrap based measure robust to the choice of normalization methods for detecting rhythmic features in high dimensional data. Front. Genet. 9:24. doi: 10.3389/fgene.2018.00024.
· R-Code, version 1.0 (578 KB)
Shyamal Peddada, Ph.D.
Proffesor and Chair, Biostatistics
130 DeSoto Street
Pittsburgh, PA 15261
Yolanda Larriba, Ph.D.
Department of Statistics and OR
Paseo de Belén, 7